When I was in my second year of college, the prof asked me to lead the choicest 3rd year project when we came back in the fall. I was giddy! The project was the North Bay municipal election. The City of North Bay had asked the Canadore College to tabulate the results of the election. And get this – voters would record their votes by sliding a computer punch card into a hand-held voting device and poking holes in the card to indicate their votes. Does this remind you of a certain election in Florida about 15 years ago? Something about Al Gore and hanging chads?
We had a bunch of debates on the numerous problems this approach raised. Hanging chads were a big one. And what about the chads that broke off in the card reader, jamming it? Card reader jams were an ongoing frustration at the college, and with the number of cards we’d be reading election night, there was little doubt we would have at least some trouble. At first we thought we should have the card reader serviced before the election, but then changed our minds, because every time the service guy messed with the mainframe, there were hiccups before things smoothed out. So instead we debated adding a second card reader. We rejected this on the grounds that the system had never run two at a time, so we could not expect flawless execution (mainframes were a little twitchy in those days).
But the worst moment came when we realized that if anyone challenged the results, we could never repeat them. Cards occasionally get destroyed in a card reader. Partially poked chads might be read as closed on the first run, and then loosen and fall off in a subsequent run, reading as open. There was no way to repeat any results we might get by re-reading the same cards.
I got a job offer just before my 3rd year started, so I never got to do the project. But I did show up for election night (another student, Paul, had been selected to run the project). Things went pretty smoothly, and thankfully, no one challenged the election results.
Lesson: Some projects have no-win components. Sometimes they’ll bite you, and sometimes you’ll get lucky.
Addendum: Late on election night, a minor flaw was apparent on the big election results monitor in the college cafeteria, where the news media and some of the candidates had gathered. According to the numbers displayed, more polls had reported than actually existed. It turned out that, for various reasons, handling all those physical voting cards made it impossible to reliably ensure a single polling station’s cards went through as a single large batch. Stray cards that eventually got read in were accidentally counted as more polling stations, when in fact they belonged to polling stations that had already been read in. I explained that to the polite gentleman I had just met standing next to me.
The next day, while quietly reading the election coverage in the North Bay Nugget, I found out he was a Nugget reporter, and had quoted me. I don’t recall him revealing his identity or asking permission.